'PI
Br. J. Pharmacol. (1992), 105, 885-892
Effects of purines
on
Macmillan Press Ltd, 1992
the longitudinal muscle of the rat colon
S.J. Bailey & 'S.M.O. Hourani Receptors and Cellular Regulation Research Group, School of Biological Sciences, University of Surrey, Guildford, Surrey GU2 5XH 1 Adenosine and adenosine 5'-triphosphate (ATP) have been reported to cause relaxation of the rat colon longitudinal muscle preparation; the purinoceptors mediating this effect were investigated by use of a series of agonists and antagonists. 2 The tissue was precontracted with carbachol (I gM), and the purines induced reversible relaxations with a potency order of 5'-N-ethylcarboxamidoadenosine (NECA) >N6-cyclopentyladenosine (CPA) = adenosine 5'-(a,-methylene) triphosphonate (AMPCPP) >adenosine= adenylyl 5'-(P,Tmethylene) disphosphonate (AMPPCP) = ATP. The PI-selective antagonist 1 ,3-dipropyl-8cyclopentylxanthine (DPCPX) (3 JLM) shifted to the right the log concentration-response curves of all these agonists except for AMPCPP, indicating that they all act via PI-purinoceptors. The order of potency of the adenosine analogues and the relatively high concentrations of the antagonist required indicated that these receptors are of the A2 subtype. The P2-selective antagonist suramin (300 liM) inhibited responses to AMPCPP, but not to the other agonists. 3 The dephosphorylation of the nucleotides was studied by high performance liquid chromatography following incubation with the longitudinal muscle preparation for up to 30 min. ATP was rapidly degraded, largely to adenosine, and AMPPCP and AMPCPP were also degraded, although more slowly, to adenosine and adenosine 5'-(a,4-methylene) diphosphonate (AMPCP) respectively. AMPCP, like AMPCPP, caused relaxations by acting on P2-purinoceptors, as it was also inhibited by suramin (300 ftM). Incubation of the tissue with adenosine deaminase abolished responses to adenosine, reduced those to ATP and AMPPCP, but had no effect on those to AMPCPP. ATP and AMPPCP therefore appear to be acting on the A2 receptors in this tissue largely via their degradation product adenosine. 4 The longitudinal muscle of the rat colon therefore contains both P.- and P2-purinoceptors, which both mediate relaxation. The P,-purinoceptors are of the A2 subtype and the P2-purinoceptors are probably of the P2Y subtype, although the rapid degradation of the nucleotides means that it is difficult to classify them with certainty. Keywords: Rat colon longitudinal muscle; purinoceptors; ATP; adenosine; ectonucleotidases; suramin
Introduction The purinoceptors which mediate the many and varied pharmacological responses to adenosine and adenine nucleotides have been clearly subdivided into PI, recognising adenosine, and P2, recognising adenosine 5'-triphosphate (ATP) (Burnstock, 1978), and selective antagonists for each of these receptor types exist. Xanthine derivatives such as theophylline and its analogues are PI-purinoceptor antagonists, having no effect on the actions of ATP in most tissues, and the trypanocidal drug suramin has recently been shown to act as a P2-purinoceptor antagonist in a number of tissues (Dunn & Blakely, 1988; Stone, 1989; Den Hertog et al., 1989a; b; Hoyle et al., 1990; Burnstock, 1990; Kennedy, 1990; Leff et al., 1990; Bailey et al., 1992). As well as acting at P2-purinoceptors, ATP is rapidly and sequentially dephosphorylated by ectonucleotidases present on the surface of cells, ultimately to adenosine, which is more slowly taken up into cells or deaminated to the inactive inosine (Pearson & Slakey, 1990). This transformation of ATP into adenosine may affect the observed responses to ATP, as not only will it reduce its effects at P2-purinoceptors but it may also result in indirect effects at P1-purinoceptors. Unfortunately, no selective inhibitors of this breakdown are available (Hourani & Chown, 1989), but the contribution of adenosine to the effects of ATP can be investigated using adenosine deaminase (ADA) to remove any adenosine produced, or inhibitors of adenosine uptake such as S-(4-nitrobenzyl)-6-thioguanosine (NBTG) to enhance the effects of adenosine.
I
Author for correspondence
P1-purinoceptors have been subdivided into Al and A2, and these can be distinguished in functional studies by the order of potency of adenosine analogues, with N6-substituted compounds such as N6-cyclopentyladenosine (CPA) being more potent than 5'-substituted compounds such as 5'-Nethylcarboxamidoadenosine (NECA) on Al receptors, but less potent on A2 receptors. In addition, the antagonist 1,3dipropyl-8-cyclopentylxanthine (DPCPX) has nanomolar affinity for Al receptors and micromolar affinity for A2 receptors, and clearly distinguishes between the two receptor subtypes in isolated tissue preparations (Collis et al., 1989; Collis, 1990; Bruns, 1990; Jacobson, 1990). In general, A2 receptors have been found to mediate the inhibitory effects of adenosine on smooth muscle, causing either relaxation of precontracted muscle or reduction in the contractile effects of other agonists, whereas Al receptors mediate contractile effects and the presynaptic inhibition of transmitter release (White, 1988; Olsson & Pearson, 1990). P2-purinoceptors on smooth muscle have also been subdivided into P2x and P2Y (Burnstock & Kennedy, 1985; Gordon, 1986) and these may be distinguished by the use of ATP analogues (Cusack & Hourani, 1990), although this subdivision is not as firmly based as that for P,-purinoceptors as no selective competitive antagonists exist, suramin being equally effective on each subtype (Dunn & Blakely, 1988; Den Hertog et al., 1989a; b; Hoyle et al., 1990; Leff et al., 1990). On P2y receptors, 2-substituted analogues such as 2-methylthioadenosine 5'-triphosphate (2-MeSATP) are more potent than ATP which is more potent than its methylene phosphonate analogues such as adenosine 5'-(a,-methylene) triphosphonate (AMPCPP) and adenylyl 5'-(P,T-methylene)
886
S.J. BAILEY & S.M.O. HOURANI
diphosphonate (AMPPCP), whereas on P2x receptors AMPPCP and AMPCPP are more potent than ATP which is equipotent with 2-MeSATP (Burnstock & Kennedy, 1985). Although the structure-activity relationships for ATP and its analogues in some tissues may be affected by the rapid degradation of ATP and its 2-substituted analogues and the relative stability of the methylene phosphonate analogues (Welford et al., 1986; 1987), the different structure-activity relationships for P2x and P2Y receptors have been confirmed by use of a series of stable isopolar phosphonate ATP analogues (Cusack et al., 1987). In addition, two nucleotide analogues, L-adenylyl 5'-(Py-methylene) diphosphonate (LAMPPCP) and adenosine 5'-(2-fluorodiphosphate) (ADP-pF) have been developed as selective agonists for P2x and P2y receptors respectively (Hourani et al., 1986; 1988). Although in general, P2x receptors mediate contraction of smooth muscle and P2y receptors mediate relaxation and this was a basis for the original subdivision (Burnstock & Kennedy, 1985), this is not a secure criterion for receptor classification and indeed in two tissues, neonatal rat duodenum and the rat colon muscularis mucosae, ATP induces contraction not relaxation via receptors whose structure-activity relationships are indicative of the P2Y subtype (Nicholls et al., 1990; Bailey & Hourani, 1990). A question which has been a matter of continuing controversy is the extent to which ATP may interact directly with P1 receptors (Stone, 1989; Bruns, 199G). There have been a number of reports of the effects of ATP being blocked by PI antagonists in some tissues, although it is not always clear whether this is a direct effect of ATP or due to its degradation to adenosine. In some cases the more stable analogue AMPPCP has also been shown to have P1 effects, which has generally been interpreted as indicating that the P1 effects of ATP itself are direct, not via its breakdown to adenosine (Dahlen & Hedqvist, 1980; Collis & Pettinger, 1982; Wiklund et al., 1985), although it has also been taken as evidence that AMPPCP is itself degraded (Moody & Burnstock, 1982), or that a third class of purinoceptor, called P3, exists at which both adenosine and adenine nucleotides are active (Shinozuka et al., 1988). This controversy may be resolved by our recent finding that in some tissues AMPPCP itself can have direct effects on P,-purinoceptors whereas ATP, 2MeSATP and AMPCPP do not, in spite of the fact that ATP is much more rapidly degraded to adenosine than is AMPPCP. In the rat colon muscularis mucosae and the rat duodenum, AMPPCP appears to act almost entirely via P1 receptors rather than by the P2y receptors which are also present, as its effects can be completely inhibited by the P1 antagonist, 8-sulphophenyltheophylline (8-SPT) in the same way as those of adenosine, whereas in the guinea-pig taenia caeci, 8-SPT causes only a roughly two fold shift in the concentration-response curve to AMPPCP, indicating that it acts largely via the P2y receptors here (Bailey & Hourani, 1990; Hourani et al., 1991). Somewhat different pA2 values were obtained for 8-SPT against AMPPCP in these tissues (approximately 5 in the taenia caeci, approximately 6 in the duodenum and colon), indicating that different P1 receptors might be present, and that AMPPCP might have selectivity for Al receptors. However, as 8-SPT has been reported not to be selective for Al receptors in isolated tissues (Collis et al., 1987) although it does show some selectivity in binding studies (Bruns et al., 1986), these small differences were not easy to interpret. Another problem is that the affinity of P1 receptors for agonists and antagonists is known to be dependent on the species used, with rat tissues in general having a higher affinity than guinea-pig tissues (Collis, 1990), which could also explain the differences we observed with 8-SPT. However, using a series of agonists and the highly A1selective antagonist DPCPX we have shown that on the rat colon muscularis mucosae the P1 receptors are of the A1 subtype whereas those on the rat duodenum are a mixture of Al and A2, with AMPPCP apparently acting on the Al population although adenosine acts on the A2 (Nicholls et
al., 1992; Bailey et al., 1992). As the P1 receptors on the taenia caeci are known to be of the A2 subtype (Burnstock et al., 1984), this would be consistent with the hypothesis that AMPPCP has selectivity for A1 receptors, although species differences could also play a role. In addition, in the rat urinary bladder in which AMPPCP causes contractions via P2x receptors, its effects were not enhanced by a concentration of DPCPX which inhibited the effect of adenosine, suggesting that it had no action on the inhibitory A2 receptors present in this tissue (Nicholls et al., 1992). To clarify this matter we wished to use a rat tissue similar to the guinea-pig taenia caeci, which relaxes to purines via P2y and A2 receptors, and to investigate whether AMPPCP would have any P1 effect in this tissue. Rather than use a vascular tissue such as the rat aorta in which there is the complication that any observed relaxation to adenosine or to ATP may be indirect, via receptors on the endothelium rather than on the smooth muscle (Rose'Meyer & Hope, 1990; Olsson & Pearson, 1990), we chose to investigate the longitudinal muscle of the rat colon. This tissue has not been widely used but has been reported to relax in response to adenosine and to ATP when precontracted with the muscarinic antagonist oxotremorine (Romano, 1981), and therefore appeared likely to be pharmacologically equivalent to the guinea-pig taenia caeci.
Methods
Pharmacological studies Male Wistar rats (150-250 g) were killed by cervical dislocation and the distal colon removed and placed in warm (320C) Tyrode buffer (ionic composition (mM): Na' 149.1, K+ 2.8, Ca2+ 1.8, Mg2+ 2.1, Cl- 147.5, H2PO4-0.3, HCO3-i1.9, glucose 5.6) pregassed with 95%02/5% CO2. The dissection of the longitudinal muscle of the colon was carried out as described by Romano (1981) and Bailey & Jordan (1984), with minor modifications. Briefly, a glass pipette, external diameter 5 mm, was placed inside the colon and the outer tubular layer of longitudinal muscle was removed by gentle rubbing with wet cotton wool, leaving a thick walled tube, the muscularis mucosae. The longitudinal muscle was suspended in a 3 ml organ bath at 32°C in gassed Tyrode solution, and contractions were recorded isometrically under a resting tension of 1 g with a Grass FT03 strain gauge and displayed on a Grass 79C polygraph. The tone of the tissue was raised with carbachol (1 gM for 3 min, which caused 50-70% of the maximal contraction) and relaxations to adenosine and adenine nucleotides were expressed as % relaxation of this carbachol-induced contraction. The longitudinal muscle was allowed to equilibrate for 90 min before control concentration-response curves to purinoceptor agonists were determined, followed by incubation with an antagonist, ADA or NBTG for 40 min before the concentration-response curves were repeated in the presence of these substances. Recovery of responses to the agonists was established following washout of these substances for up to 40 min; 12-20 min were allowed between doses of the purinoceptor agonists, which were left in contact with the tissue for 2-5 min until a maximal relaxation had been observed.
Degradation studies Segments of rat colon longitudinal muscle (approximately 30 mg wet weight) were suspended in 3 ml organ baths as described for the pharmacological studies. Following preincubation for 90-120 min with frequent washing, each was exposed to ATP, AMPPCP and AMPCPP (100 jIM at 30 min intervals), and samples of the bathing medium were taken at 2, 5, 10 and 20 min and frozen for later analysis by high performance liquid chromatography as described by Welford et al. (1986, 1987).
PURINES AND THE RAT COLON LONGITUDINAL MUSCLE
887
Materials ATP, AMPPCP, AMPCPP, AMPCP, adenosine, ADA (Type VI) and NBTG were obtained from Sigma UK. Ltd, DPCPX, 2-MeSATP, CPA and NECA were obtained from Research Biochemicals, and suramin was a generous gift from Bayer, UK. Other purine analogues were kindly provided by Dr Noel J. Cusack, Whitby Research Inc, Richmond, Virginia, U.S.A. CPA (10 mM) was dissolved in 20% ethanol, DPCPX (1 mM) was dissolved in 2% aqueous dimethylsulphoxide (DMSO) containing 6 mM NaOH and NBTG (50 mM) was dissolved in DMSO. After dilution corresponding to the final bath concentration of the substances used these solvents had no effect on the responses of the tissue. The ADA solution was supplied in 50% glycerol0.01 M potassium phosphate and was diluted in distilled water to give a stock solution of 200 units ml-'.
t~0 40-I co
0 0.1
1
10
100
Purine concentration (>M)
Results Adenosine, ATP, AMPPCP, AMPCPP, NECA and CPA each relaxed the carbachol-contracted rat colon longitudinal muscle, and the order of potency was NECA> CPA = AMPCPP> adenosine = AMPPCP = ATP (Figure 1). DPCPX
d
a
100
Figure I Relaxation of the rat colon longitudinal muscle by adenosine (@), NECA (A), CPA (-), ATP (0), AMPPCP (') or AMPCPP (V). Each point is the mean of at least 15 determinations, and the error bars, which never exceeded 5%, have been omitted for clarity. For abbreviations, see text.
100 -
-
50 -
50 -
O-
0- l
10
1
100
100
ATP concentration (>.M)
Adenosine concentration (>M) 1
b
100
100 -i
-
50-
(a
x
la
W-J
0.1
1
1
10
100
AMPPCP concentration (>M)
NECA concentration (>M) f
C
100 -
100 -
50 -
50-
0-
10
I
I
I
1
10
100
CPA concentration (>±M)
0-
10
100
AMPCPP concentration (>M)
Figure 2 Relaxation of the rat colon longitudinal muscle by (a) adenosine, (b) NECA, (c) CPA, (d) ATP, (e) AMPPCP or (f) AMPCPP alone (@) or in the presence of DPCPX (3 JiM) (0). Each point is the mean of at least 4 determinations and the vertical bars show s.e.mean. For abbreviations, see text.
888
ST BAILEY & S.M.O. HOURANI
(3 gM) inhibited the responses to NECA, CPA, adenosine, AMPPCP and ATP, but did not affect the relaxations induced by AMPCPP (Figure 2), whereas suramin (3001MM) inhibited the responses to AMPCPP but not those of the other agonists (Figure 3). DPCPX (1 gM) caused smaller shifts to the right of the log concentration-response curves to all the agonists except for AMPCPP (results not shown), but Schild analysis was not performed as concentrations of DPCPX above 3pM could not be achieved. L-AMPPCP (100 gM) and 2-MeSATP (100m M) were almost inactive, ADP-P-F(l00 Mm) induced a relaxation similar to that of ATP (100 IM), while adenosine 5'-(Q,-methylene) diphosphonate (AMPCP) (1001MM) induced a relaxation similar to that of AMPCPP (100m M) (Figure 4). Representative traces showing the relaxations induced by the agonists are shown in Figure 4. The responses to AMPCP were inhibited by suramin (300 gM) in a similar manner to those of AMPCPP (results not shown). Incubation of the tissues with ADA (2 units ml-') abolished the relaxations induced by adenosine and inhibited those induced by ATP and AMPPCP, but had no effect on those induced by AMPCPP (Figure 5). Incubation of the tissues with NBTG (50 gM) potentiated responses to adenosine, causing a roughly three fold shift to the left of the log concentration-response curve, and also potentiated responses to AMPCPP but to a somewhat lesser extent (results not shown).
Degradation studies Dephosphorylation of ATP by the rat longitudinal muscle preparation was rapid, with approximately 80% being degraded, largely to adenosine and AMP, during 30 min incubation. AMPPCP was degraded more slowly to adenosine, with approximately 20% conversion during 30 min incubation, and AMPCPP was degraded to AMPCP, with approximately 30% conversion over 30 min (Figure 6).
Discussion These results show that the outer longitudinal muscle layer of the rat colon contains P1- and P2-purinoceptors which both mediate relaxation, and that the responses of this tissue to purines are qualitatively similar to the responses seen in the guinea-pig taenia caeci, the longitudinal muscle of the guinea-pig caecum. For the Pl-purinoceptor in the longitudinal muscle the potency order of agonists is NECA> CPA> adenosine, and DPCPX has an apparent dissociation constant in the micromolar range, suggesting that the receptor is of the A2 subtype (Bruns, 1990; Collis, 1990; Daly, 1990). DPCPX (3 gM) caused a somewhat smaller shift to the right of the log concentration-response curve to adenosine compared with NECA and CPA, which may suggest that it also interfered with adenosine uptake. In general the postd
a
100-
100
50
50-
-
0.
0 *1
10
100
1
100
ATP concentration (>tM)
Adenosine concentration (>xM) 100 -1
e 100-
50 -
50-
b
10
R C
-
0
ur
(U
CC
0-
0-
0.1
1
1
10
100
AMPPCP concentration (>tM)
NECA concentration ({M) c
10
f
100-
100-
50-
50-
0-
0 10
100
CPA concentration (>M)
1
10
100
AMPCPP concentration (>M )
Figure 3 Relaxation of the rat colon longitudinal muscle by (a) adenosine, (b) NECA, (c) CPA, (d) ATP, (e) AMPPCP or (f) AMPCPP alone (@) or in the presence of suramin (300 IpM) (0). Each point is the mean of at least 4 determinations and the vertical bars show s.e.mean. For abbreviations, see text.
PURINES AND THE RAT COLON LONGITUDINAL MUSCLE
C Adenosine
C NECA
C CPA
J1 g
/t
4 min
r
C ATP
C AMPPCP
C AMPCPP
C AMPCP
C 2t C 2-MeSATP
C ADP-3-F
C L-AMPPCP
Figure 4 Representative traces showing relaxations of the rat colon longitudinal muscle to purines after precontraction of the tissue with carbachol (1 #tM) (C). The concentration of purine added in each case was 100 IAM, except for NECA which was 10 I1M. For abbreviations used in figure, see text.
100-
50 -
50-
0-
0C
synaptic inhibitory effects of adenosine in smooth muscle including the guinea-pig taenia caeci are mediated by A2 receptors, so the finding that the adenosine receptor here is of the A2 subtype is consistent with results in other tissues (White, 1988; Stone, 1989; Kennedy, 1990; Olsson & Pearson, 1990). In those tissues in which adenosine causes contraction, such as the renal vasculature (Kenakin & Pike, 1987) and the guinea-pig myometrium (Smith et al., 1988), this appears to be via Al receptors, and this is also the case in the rat colon muscularis mucosae (Bailey et al., 1992). It has also been reported that the guinea-pig aorta and trachea each possess both Al and A2 receptors which mediate contraction and relaxation respectively, although for adenosine the A2-mediated relaxation is dominant (Farmer et al., 1988; Stoggall & Shaw, 1990). The existence in the rat colon of excitatory Al receptors (on the muscularis mucosae) and inhibitory A2 receptors (on the longitudinal muscle) is therefore consistent with this general pattern, although the two receptor types are clearly at anatomically distinct locations within the colon, a possibility which has not been investigated in the other tissues. The rat colon longitudinal muscle also possesses P2purinoceptors, as AMPCPP caused a relaxation which was inhibited by the P2-purinoceptor antagonist suramin but not by DPCPX. The actions of ATP and AMPPCP appeared however to be mediated via the A2 receptor, as they were not inhibited by suramin but were inhibited by DPCPX at the same concentration as was adenosine. It seems likely that this A2 effect was at least partially indirect and a result of the degradation of ATP and AMPPCP to adenosine, as the responses to ATP and AMPPCP were reduced by ADA, which had no effect on the responses to AMPCPP. The adenosine uptake inhibitor NBTG potentiated responses to adenosine as expected, but also potentiated responses to AMPCPP which does not act on A2 receptors here, and therefore was not useful for resolving the question as to whether the actions of ATP and AMPPCP were direct or via adenosine. The potentiation by NBTG of responses to AMP-
b 100 -j
a
1
0
100
1
100 Adenosine concentration (j±M) 10
ATP concentration (jiM)
x
cc
d
c
100-
100 -,
50-
50-
0-
01
10
100
AMPPCP concentrtion (gM)
889
1
10
100
AMPCPP concentration
(gM)
Figure 5 Relaxation of the rat colon longitudinal muscle by (a) adenosine, (b) ATP, (c) AMPPCP or (d) AMPCPP alone (0) or in the presence of adenosine deaminase (2 units ml- 1) (0). Each point is the mean of at least 5 determinations and the vertical bars show s.e.mean. For abbreviations, see text.
890
S.J. BAILEY & S.M.O. HOURANI 100
a -k
b
100 -
0
ci
C1)
50
C) 0 C.)
a1)
C)
CL
0
T
10
0
20
7 30
100 c
50-
0
the 0
10
I
I
20
30
Incubation time (min)
Figure 6 Degradation of purines by the rat colon longitudinal muscle preparation. Removal of (a) ATP, (b) AMPPCP or (c) AMPCPP (0) and appearance of ADP (A), AMP (0), adenosine (V) or AMPCP (0). Each point is the mean of at least 4 determinations and the vertical bars show s.e. mean. For abbreviations, see text.
CPP was presumably due to increased endogenous adenosine levels enhancing all relaxant responses, but this was not investigated further. It is probable that AMPCPP retained its
P2-purinoceptor activity even though it is also significantly degraded, to at least the same extent as AMPPCP, because its breakdown product was not adenosine but AMPCP, which was inhibited by suramin and therefore acts on P2purinoceptors. These findings are in contrast to our results with the rat colon muscularis mucosae and the rat duodenum, where it is clear that despite its rapid degrada-
tion, ATP has little if any P1-purinoceptor activity whereas AMPPCP acts almost entirely and directly via P1purinoceptors (Bailey & Hourani, 1990; Hourani et al., 1991). Unfortunately the uncertainty regarding the direct A2 receptor activity of AMPPCP itself in the longitudinal muscle means that the use of this tissue has not resolved our initial hypothesis that AMPPCP might have Al-selective PIpurinoceptor activity, as had been suggested by our previous studies (Bailey & Hourani, 1990; Nicholls et al., 1990; Hourani et al., 1991; Bailey et al., 1992; Nicholls et al., 1992). Although the P2-purinoceptor which mediates relaxation in the guinea-pig taenia caeci is clearly of the P2Y subtype, with an order of potency of 2-MeSATP> ATP> AMPCPP (Burnstock et al., 1984; Burnstock & Kennedy, 1985), the P2purinoceptor in the rat colon longitudinal muscle is not so easy to define by use of these agonists as ATP itself had no P2-purinoceptor activity in this tissue and 2-MeSATP was almost inactive. The weak activity of 2-MeSATP may be due to its rapid breakdown to 2-methylthioadenosine, which is somewhat less active than adenosine on the A2 receptors in the guinea-pig taenia caeci (Satchell & Maguire, 1975) and is therefore likely to be less active in the longitudinal muscle preparation. Indeed the very rapid degradation of the nucleotides, including the normally resistant AMPPCP and AMPCPP, by this preparation makes analysis of the structure-activity relationships of the P2-purinoceptor very difficult. The response mediated by this receptor is a relaxation which is qualitatively similar to that seen in the guineapig taenia caeci, which might suggest involvement of a P2ypurinoceptor, but this is not a reliable method for receptor classification, particularly in view of the existence of P2Ypurinoceptors which mediate contraction on the muscularis mucosae and the neonatal rat duodenum (Bailey & Hourani, 1990; Nicholls et al., 1990). The lack of activity of LAMPPCP and the activity of ADP-P-F in the longitudinal muscle would appear to confirm that the P2-purinoceptor here is of the P2y subtype, as L-AMPPCP and ADP-13-F are selective agonists at P2X- and P2Y-purinoceptors respectively (Hourani et al., 1986; 1988). However, the same argument could in theory apply to L-AMPPCP as to 2-MeSATP, as its potential breakdown product, L-adenosine, is inactive on PIpurinoceptors (Brown et al., 1982), and similarly ADP-P-F could be acting via its probable breakdown product, adenosine, so this evidence cannot be regarded as conclusive. What is needed, both to define the P2-purinoceptor in this tissue and to resolve the question of whether ATP and AMPPCP act directly or indirectly on the P,-purinoceptor, is a selective inhibitor of the ectonucleotidases which are responsible for the breakdown. Notwithstanding these uncertainties, it is clear from the work presented here together with our previous work (Bailey & Hourani, 1990; Bailey et al., 1992), that the two preparations from the rat colon, the muscularis mucosae and the longitudinal muscle, have different P,-purinoceptor subtypes, Al mediating contraction and A2 mediating relaxation, and that they also possess P2 purinoceptors which mediate the same effect in each case as the P1 purinoceptors. The physiological significance of these receptors is unknown, but probably reflects the widespread roles of purines as transmitters and modulators in the autonomic nervous system. The nature of the P2-purinoceptor on the longitudinal muscle is still undefined, but it is likely to be of the P2Y subtype as in the muscularis mucosae, although the response observed is different in the two preparations. In the longitudinal muscle preparation the breakdown of nucleotides is extremely rapid, and is considerably faster than that previously observed in the muscularis mucosae in spite of the fact that the longitudinal muscle is a much smaller tissue, with a wet weight of about 30 mg compared to around 200 mg for the muscularis mucosae. This very rapid degradation undoubtedly affects the responses to the nucleotides, and this makes the study of the pharmacology of the P2-purinoceptors in this
PURINES AND THE RAT COLON LONGITUDINAL MUSCLE
tissue somewhat problematical, and shows clearly the care which must be taken when attempting to use nucleotide agonist potencies to define receptor subtypes.
891
We thank the Medical Research Council for support for S.J.B. (G8723473N) and Dr Noel Cusack for supplying analogues.
References
BAILEY, S.J. & JORDAN, C.C. (1984). A study of [D-Pro2, D-Phe7 D-Trpl-substance P and (D-Trp7f9l-substance P as tachykinin partial agonists in the rat colon. Br. J. Pharmacol., 82, 441-451. BAILEY, S.J., HICKMAN, D. & HOURANI, S.M.O. (1992). Characterisation of the PI-purinoceptors mediating contraction of the rat colon muscularis mucosae. Br. J. Pharmacol., 105, 400-404. BAILEY, S.J. & HOURANI, S.M.O. (1990). A study of the purinoceptors mediating contraction in the rat colon. Br. J. Pharmacol., 100, 753-756. BROWN, C., BURNSTOCK, G., CUSACK, N.J., MEGHJI, P. & MOODY,
C.J. (1982). Evidence for the stereospecificity of the PIpurinoceptor. Br. J. Pharmacol., 75, 101-107. BRUNS, R.F. (1990). Adenosine receptors: roles and pharmacology.
Ann N.Y. Acad. Sci., 603, 211-226. BRUNS, R.F., LU, G.H. & PUGSLEY, T.A. (1986). Characterization of the A2 adenosine receptor labeled by [3H]NECA in rat striatal membranes. Mol. Pharmacol., 29, 331-346. BURNSTOCK, G. (1978). A basis for distinguishing two types of purinergic receptor. In Cell and Membrane Receptors for Drugs and Hormones: a Multidisciplinary Approach. ed. Bolis, L. & Straub, R.W. pp. 107-118. New York: Raven Press. BURNSTOCK, G. (1990). Classification and characterization of purinoceptors. In Purines in Cellular Signalling. ed. Jacobson, K.A., Daly, J.W. & Manganiello, V., pp. 241-253. New York:
Springer-Verlag.
BURNSTOCK, G., HILLS, J.M. & HOYLE, C.H.V. (1984). Evidence that the PI-purinoceptor in the guinea-pig taenia coli is an A2 subtype. Br. J. Pharmacol., 81, 533-541. BURNSTOCK, G. & KENNEDY, C. (1985). Is there a basis for distinguishing two types of P2-purinoceptor? Gen. Pharmacol., 16, 433-440. COLLIS, M.G. (1990). Adenosine receptor sub-types in isolated tissues: antagonist studies. In Purines in Cellular Signalling. ed. Jacobson, K.A., Daly, J.W. & Manganiello, V. pp. 48-53. New York: Springer-Verlag. COLLIS, M.G., JACOBSON, K.A. & TOMKINS, D.M. (1987). Apparent affinity of some 8-phenyl-substituted xanthines at adenosine receptors in guinea-pig aorta and atria. Br. J. Pharmacol., 92, 69-75. COLLIS, M.G. & PETTINGER, S.J. (1982). Can ATP stimulate PIreceptors in guinea-pig atrium without conversion to adenosine? Eur. J. Pharmacol., 81, 521-529. COLLIS, M.G., STOGGALL, S.M. & MARTIN, F.M. (1989). Apparent affinity of 1,3-dipropyl-8-cyclopentylxanthine for adenosine Al and A2 receptors in isolated tissues from guinea-pigs. Br. J. Pharmacol., 97, 1274-1278. CUSACK, N.J. & HOURANI, S.M.O. (1990). Subtypes of P2purinoceptors. Studies using analogues of ATP. Ann. N.Y. Acad. Sci., 603, 172-181. CUSACK, N.J., HOURANI, S.M.O., LOIZOU, G.D. & WELFORD, L.A. (1987). Pharmacological effects of isopolar phosphonate analogues of ATP on P2-purinoceptors in guinea-pig taenia coli and urinary bladder. Br. J. Pharmacol., 90, 791-795. DAHLEN, S.-E. & HEDQVIST, P. (1980). ATP, P,y-methylene-ATP, and adenosine inhibit non-cholinergic non-adrenergic transmission in rat urinary bladder. Acta Physiol, Scand., 109, 137-142. DALY, J.W. (1990). Adenosine agonists and antagonists. In Purines in Cellular Signalling. ed. Jacobson, K.A., Daly, J.W. & Manganiello, V. pp. 3-12. New York: Springer-Verlag. DEN HERTOG, A., NELEMANS, A. & VAN DEN AKKER, J. (1989a). The inhibitory action of suramin on the P2-purinoceptor in smooth muscle cells of guinea-pig taenia caeci. Eur. J. Pharmacol., 166, 531-534. DEN HERTOG, A., VAN DEN AKKER, J. & NELEMANS, A. (1989b). Suramin and the inhibitory junction potential in taenia caeci of the guinea-pig. Eur. J. Pharmacol., 173, 207-209. DUNN, P.M. & BLAKELY, A.G.H. (1988). Suramin: a reversible P2purinoceptor antagonist in the niouse vas deferens. Br. J. Pharmacol., 93, 243-245. FARMER, S.G., CANNING, B.J. & WILKINS, D.E. (1988). Adenosine receptor-mediated contraction and relaxation of guinea-pig isolated tracheal smooth muscle: effects of adenosine antagonists. Br. J. Pharmacol., 95, 371-378.
GORDON, J.L. (1986). Extracellular ATP: effects, sources and fate. Biochem. J., 233, 309-319. HOURANI, S.M.O., BAILEY, S.J., NICHOLLS, J. & KITCHEN, I. (1991).
Direct effects of adenylyl 5'-(P,y-methylene) diphosphonate, a stable ATP analogue, on relaxant PI-purinoceptors in smooth muscle. Br. J. Pharmacol., 104, 685-690. HOURANI, S.M.O. & CHOWN, J.A. (1989). The effects of some possible inhibitors of ectonucleotidases on the breakdown and pharmacological effects of ATP in the guinea-pig urinary bladder. Gen. Pharmacol., 20, 413-416. HOURANI, S.M.O., LOIZOU, G.D. & CUSACK, N.J. (1986). Pharmacological effects of L-AMP-PCP on ATP receptors in smooth muscle. Eur. J. Pharmacol., 131, 99-103. HOURANI, S.M.O., WELFORD, L.A., LOIZOU, G.D. & CUSACK, N.J.
(1988). Adenosine 5'-(2-fluorodiphosphate) is a selective agonist at P2-purinoceptors mediating relaxation of smooth muscle. Eur. J. Pharmacol., 147, 131-136. HOYLE, C.H.V., KNIGHT, G.E. & BURNSTOCK, G. (1990). Suramin antagonizes responses to P2-purinoceptor agonists and purinergic nerve stimulation in the guinea-pig urinary bladder and taenia coli. Br. J. Pharmacol., 99, 617-621. JACOBSON, K.A. (1990). Molecular probes for adenosine receptors. In Purines in Cellular Signalling. ed. Jacobson, K.A., Daly, J.W. & Manganiello, V., pp. 54-64. New York: Springer-Verlag. KENAKIN, T.P. & PIKE, N.B. (1987). An in vitro analysis of purinemediated renal vasoconstriction in rat isolated kidney. Br. J. Pharmacol., 90, 373-381. KENNEDY, C. (1990). PI- and P2-purinoceptor subtypes - an update. Arch. Int. Pharmacodyn., 303, 30-50. LEFF, P., WOOD, B.E. & O'CONNOR, S.E. (1990). Suramin is a slowlyequilibrating but competitive antagonist at P2X-receptors in the rabbit isolated ear artery. Br. J. Pharmacol., 101, 645-649. MOODY, C.J. & BURNSTOCK, G. (1982). Evidence for the presence of P1-purinoceptors on cholinergic nerve terminals in the guinea-pig ileum. Eur. J. Pharmacol., 77, 1-9. NICHOLLS, J., HOURANI, S.M.O. & KITCHEN, I. (1990).- The ontogeny of purinoceptors in rat urinary bladder and duodenum. Br. J. Pharmacol., 100, 874-878. NICHOLLS, J., HOURANI, S.M.O. & KITCHEN, I. (1992). Characterization of P1-purinoceptors on rat duodenum and urinary bladder. Br. J. Pharmacol., (in press). OLSSON, R.A. & PEARSON, J.D. (1990). Cardiovascular purinoceptors. Pharmacol. Rev., 70, 761-845. PEARSON, J.D. & SLAKEY, L.L. (1990). Intracellular and extracellular metabolism of adenosine and adenine nucleotides. In Purines in Cellular Signalling. ed. Jacobson, K.A., Daly, J.W. & Manganiello, V. pp. 13-19. New York: Springer-Velag. ROMANO, C. (1981). Nicotine action on rat colon. J. Pharmacol. Exp. Ther., 217, 828-833. ROSE'MEYER, R.B. & HOPE, W. (1990). Evidence that A2 purinoceptors are involved in endothelium-dependent relaxation of the rat thoracic aorta. Br. J. Pharmacol., 100, 576-580. SATCHELL, D.G. & MAGUIRE, M.H. (1975). Inhibitory effects of adenine nucleotide analogs on the isolated guinea-pig taenia coli. J. Pharmacol. Exp. Ther., 195, 540-548. SHINOZUKA, K., BJUR, R.A. & WESTFALL, D.P. (1988). Characterisation of prejunctional purinoceptors on adrenergic nerves of the rat caudal artery. Naunyn-Schmiedebergs Arch. Pharmacol., 338, 221-227. SMITH, M.A., BUXTON, I.L.O. & WESTFALL, D.P. (1988). Pharmacological classification of receptors for adenyl purines in guinea-pig myometrium. J. Pharmacol. Exp. Ther., 247, 10591063. STOGGALL, S.M. & SHAW, J.S. (1990). The coexistence of adenosine Al and A2 receptors in guinea-pig aorta. Eur. J. Pharmacol., 190, 329-335. STONE, T.W. (1989). Purine receptors and their pharmacological roles. Adv. Drug Res., 18, 291-429. WELFORD, L.A., CUSACK, N.J. & HOURANI, S.M.O. (1986). ATP analogues and the guinea-pig taenia coli: a comparison of the structure-activity relationships of ectonucleotidases with those of the P2-purinoceptor. Eur. J. Pharmacol., 129, 217-224.
892
S.J. BAILEY & S.M.O. HOURANI
WELFORD, L.A., CUSACK, N.J. & HOURANI, S.M.O. (1987). The
structure-activity relationships of ectonucleotidases and of excitatory P2-purinoceptors: evidence that dephosphorylation of ATP analogues reduces pharmacological potency. Eur. J. Pharmacol., 141, 123-130. WHITE, T.D. (1988). Role of adenine compounds in autonomic neurotransmission. Pharmacol. Ther., 38, 129-168.
WIKLUND, N.P., GUSTAFSSON, L.E. & LUNDIN, J. (1985). Pre- and post-junctional modulation of cholinergic neuroeffector transmission by adenine nucleotides. Experiments with agonist and antagonist. Acta Physiol. Scand., 125, 681-691. (Received October 3, 1991 Revised December 6, 1991 Accepted December 18, 1991)
Br. J. Pharmacol. (1992), 105, 885-892
Effects of purines
on
Macmillan Press Ltd, 1992
the longitudinal muscle of the rat colon
S.J. Bailey & 'S.M.O. Hourani Receptors and Cellular Regulation Research Group, School of Biological Sciences, University of Surrey, Guildford, Surrey GU2 5XH 1 Adenosine and adenosine 5'-triphosphate (ATP) have been reported to cause relaxation of the rat colon longitudinal muscle preparation; the purinoceptors mediating this effect were investigated by use of a series of agonists and antagonists. 2 The tissue was precontracted with carbachol (I gM), and the purines induced reversible relaxations with a potency order of 5'-N-ethylcarboxamidoadenosine (NECA) >N6-cyclopentyladenosine (CPA) = adenosine 5'-(a,-methylene) triphosphonate (AMPCPP) >adenosine= adenylyl 5'-(P,Tmethylene) disphosphonate (AMPPCP) = ATP. The PI-selective antagonist 1 ,3-dipropyl-8cyclopentylxanthine (DPCPX) (3 JLM) shifted to the right the log concentration-response curves of all these agonists except for AMPCPP, indicating that they all act via PI-purinoceptors. The order of potency of the adenosine analogues and the relatively high concentrations of the antagonist required indicated that these receptors are of the A2 subtype. The P2-selective antagonist suramin (300 liM) inhibited responses to AMPCPP, but not to the other agonists. 3 The dephosphorylation of the nucleotides was studied by high performance liquid chromatography following incubation with the longitudinal muscle preparation for up to 30 min. ATP was rapidly degraded, largely to adenosine, and AMPPCP and AMPCPP were also degraded, although more slowly, to adenosine and adenosine 5'-(a,4-methylene) diphosphonate (AMPCP) respectively. AMPCP, like AMPCPP, caused relaxations by acting on P2-purinoceptors, as it was also inhibited by suramin (300 ftM). Incubation of the tissue with adenosine deaminase abolished responses to adenosine, reduced those to ATP and AMPPCP, but had no effect on those to AMPCPP. ATP and AMPPCP therefore appear to be acting on the A2 receptors in this tissue largely via their degradation product adenosine. 4 The longitudinal muscle of the rat colon therefore contains both P.- and P2-purinoceptors, which both mediate relaxation. The P,-purinoceptors are of the A2 subtype and the P2-purinoceptors are probably of the P2Y subtype, although the rapid degradation of the nucleotides means that it is difficult to classify them with certainty. Keywords: Rat colon longitudinal muscle; purinoceptors; ATP; adenosine; ectonucleotidases; suramin
Introduction The purinoceptors which mediate the many and varied pharmacological responses to adenosine and adenine nucleotides have been clearly subdivided into PI, recognising adenosine, and P2, recognising adenosine 5'-triphosphate (ATP) (Burnstock, 1978), and selective antagonists for each of these receptor types exist. Xanthine derivatives such as theophylline and its analogues are PI-purinoceptor antagonists, having no effect on the actions of ATP in most tissues, and the trypanocidal drug suramin has recently been shown to act as a P2-purinoceptor antagonist in a number of tissues (Dunn & Blakely, 1988; Stone, 1989; Den Hertog et al., 1989a; b; Hoyle et al., 1990; Burnstock, 1990; Kennedy, 1990; Leff et al., 1990; Bailey et al., 1992). As well as acting at P2-purinoceptors, ATP is rapidly and sequentially dephosphorylated by ectonucleotidases present on the surface of cells, ultimately to adenosine, which is more slowly taken up into cells or deaminated to the inactive inosine (Pearson & Slakey, 1990). This transformation of ATP into adenosine may affect the observed responses to ATP, as not only will it reduce its effects at P2-purinoceptors but it may also result in indirect effects at P1-purinoceptors. Unfortunately, no selective inhibitors of this breakdown are available (Hourani & Chown, 1989), but the contribution of adenosine to the effects of ATP can be investigated using adenosine deaminase (ADA) to remove any adenosine produced, or inhibitors of adenosine uptake such as S-(4-nitrobenzyl)-6-thioguanosine (NBTG) to enhance the effects of adenosine.
I
Author for correspondence
P1-purinoceptors have been subdivided into Al and A2, and these can be distinguished in functional studies by the order of potency of adenosine analogues, with N6-substituted compounds such as N6-cyclopentyladenosine (CPA) being more potent than 5'-substituted compounds such as 5'-Nethylcarboxamidoadenosine (NECA) on Al receptors, but less potent on A2 receptors. In addition, the antagonist 1,3dipropyl-8-cyclopentylxanthine (DPCPX) has nanomolar affinity for Al receptors and micromolar affinity for A2 receptors, and clearly distinguishes between the two receptor subtypes in isolated tissue preparations (Collis et al., 1989; Collis, 1990; Bruns, 1990; Jacobson, 1990). In general, A2 receptors have been found to mediate the inhibitory effects of adenosine on smooth muscle, causing either relaxation of precontracted muscle or reduction in the contractile effects of other agonists, whereas Al receptors mediate contractile effects and the presynaptic inhibition of transmitter release (White, 1988; Olsson & Pearson, 1990). P2-purinoceptors on smooth muscle have also been subdivided into P2x and P2Y (Burnstock & Kennedy, 1985; Gordon, 1986) and these may be distinguished by the use of ATP analogues (Cusack & Hourani, 1990), although this subdivision is not as firmly based as that for P,-purinoceptors as no selective competitive antagonists exist, suramin being equally effective on each subtype (Dunn & Blakely, 1988; Den Hertog et al., 1989a; b; Hoyle et al., 1990; Leff et al., 1990). On P2y receptors, 2-substituted analogues such as 2-methylthioadenosine 5'-triphosphate (2-MeSATP) are more potent than ATP which is more potent than its methylene phosphonate analogues such as adenosine 5'-(a,-methylene) triphosphonate (AMPCPP) and adenylyl 5'-(P,T-methylene)
886
S.J. BAILEY & S.M.O. HOURANI
diphosphonate (AMPPCP), whereas on P2x receptors AMPPCP and AMPCPP are more potent than ATP which is equipotent with 2-MeSATP (Burnstock & Kennedy, 1985). Although the structure-activity relationships for ATP and its analogues in some tissues may be affected by the rapid degradation of ATP and its 2-substituted analogues and the relative stability of the methylene phosphonate analogues (Welford et al., 1986; 1987), the different structure-activity relationships for P2x and P2Y receptors have been confirmed by use of a series of stable isopolar phosphonate ATP analogues (Cusack et al., 1987). In addition, two nucleotide analogues, L-adenylyl 5'-(Py-methylene) diphosphonate (LAMPPCP) and adenosine 5'-(2-fluorodiphosphate) (ADP-pF) have been developed as selective agonists for P2x and P2y receptors respectively (Hourani et al., 1986; 1988). Although in general, P2x receptors mediate contraction of smooth muscle and P2y receptors mediate relaxation and this was a basis for the original subdivision (Burnstock & Kennedy, 1985), this is not a secure criterion for receptor classification and indeed in two tissues, neonatal rat duodenum and the rat colon muscularis mucosae, ATP induces contraction not relaxation via receptors whose structure-activity relationships are indicative of the P2Y subtype (Nicholls et al., 1990; Bailey & Hourani, 1990). A question which has been a matter of continuing controversy is the extent to which ATP may interact directly with P1 receptors (Stone, 1989; Bruns, 199G). There have been a number of reports of the effects of ATP being blocked by PI antagonists in some tissues, although it is not always clear whether this is a direct effect of ATP or due to its degradation to adenosine. In some cases the more stable analogue AMPPCP has also been shown to have P1 effects, which has generally been interpreted as indicating that the P1 effects of ATP itself are direct, not via its breakdown to adenosine (Dahlen & Hedqvist, 1980; Collis & Pettinger, 1982; Wiklund et al., 1985), although it has also been taken as evidence that AMPPCP is itself degraded (Moody & Burnstock, 1982), or that a third class of purinoceptor, called P3, exists at which both adenosine and adenine nucleotides are active (Shinozuka et al., 1988). This controversy may be resolved by our recent finding that in some tissues AMPPCP itself can have direct effects on P,-purinoceptors whereas ATP, 2MeSATP and AMPCPP do not, in spite of the fact that ATP is much more rapidly degraded to adenosine than is AMPPCP. In the rat colon muscularis mucosae and the rat duodenum, AMPPCP appears to act almost entirely via P1 receptors rather than by the P2y receptors which are also present, as its effects can be completely inhibited by the P1 antagonist, 8-sulphophenyltheophylline (8-SPT) in the same way as those of adenosine, whereas in the guinea-pig taenia caeci, 8-SPT causes only a roughly two fold shift in the concentration-response curve to AMPPCP, indicating that it acts largely via the P2y receptors here (Bailey & Hourani, 1990; Hourani et al., 1991). Somewhat different pA2 values were obtained for 8-SPT against AMPPCP in these tissues (approximately 5 in the taenia caeci, approximately 6 in the duodenum and colon), indicating that different P1 receptors might be present, and that AMPPCP might have selectivity for Al receptors. However, as 8-SPT has been reported not to be selective for Al receptors in isolated tissues (Collis et al., 1987) although it does show some selectivity in binding studies (Bruns et al., 1986), these small differences were not easy to interpret. Another problem is that the affinity of P1 receptors for agonists and antagonists is known to be dependent on the species used, with rat tissues in general having a higher affinity than guinea-pig tissues (Collis, 1990), which could also explain the differences we observed with 8-SPT. However, using a series of agonists and the highly A1selective antagonist DPCPX we have shown that on the rat colon muscularis mucosae the P1 receptors are of the A1 subtype whereas those on the rat duodenum are a mixture of Al and A2, with AMPPCP apparently acting on the Al population although adenosine acts on the A2 (Nicholls et
al., 1992; Bailey et al., 1992). As the P1 receptors on the taenia caeci are known to be of the A2 subtype (Burnstock et al., 1984), this would be consistent with the hypothesis that AMPPCP has selectivity for A1 receptors, although species differences could also play a role. In addition, in the rat urinary bladder in which AMPPCP causes contractions via P2x receptors, its effects were not enhanced by a concentration of DPCPX which inhibited the effect of adenosine, suggesting that it had no action on the inhibitory A2 receptors present in this tissue (Nicholls et al., 1992). To clarify this matter we wished to use a rat tissue similar to the guinea-pig taenia caeci, which relaxes to purines via P2y and A2 receptors, and to investigate whether AMPPCP would have any P1 effect in this tissue. Rather than use a vascular tissue such as the rat aorta in which there is the complication that any observed relaxation to adenosine or to ATP may be indirect, via receptors on the endothelium rather than on the smooth muscle (Rose'Meyer & Hope, 1990; Olsson & Pearson, 1990), we chose to investigate the longitudinal muscle of the rat colon. This tissue has not been widely used but has been reported to relax in response to adenosine and to ATP when precontracted with the muscarinic antagonist oxotremorine (Romano, 1981), and therefore appeared likely to be pharmacologically equivalent to the guinea-pig taenia caeci.
Methods
Pharmacological studies Male Wistar rats (150-250 g) were killed by cervical dislocation and the distal colon removed and placed in warm (320C) Tyrode buffer (ionic composition (mM): Na' 149.1, K+ 2.8, Ca2+ 1.8, Mg2+ 2.1, Cl- 147.5, H2PO4-0.3, HCO3-i1.9, glucose 5.6) pregassed with 95%02/5% CO2. The dissection of the longitudinal muscle of the colon was carried out as described by Romano (1981) and Bailey & Jordan (1984), with minor modifications. Briefly, a glass pipette, external diameter 5 mm, was placed inside the colon and the outer tubular layer of longitudinal muscle was removed by gentle rubbing with wet cotton wool, leaving a thick walled tube, the muscularis mucosae. The longitudinal muscle was suspended in a 3 ml organ bath at 32°C in gassed Tyrode solution, and contractions were recorded isometrically under a resting tension of 1 g with a Grass FT03 strain gauge and displayed on a Grass 79C polygraph. The tone of the tissue was raised with carbachol (1 gM for 3 min, which caused 50-70% of the maximal contraction) and relaxations to adenosine and adenine nucleotides were expressed as % relaxation of this carbachol-induced contraction. The longitudinal muscle was allowed to equilibrate for 90 min before control concentration-response curves to purinoceptor agonists were determined, followed by incubation with an antagonist, ADA or NBTG for 40 min before the concentration-response curves were repeated in the presence of these substances. Recovery of responses to the agonists was established following washout of these substances for up to 40 min; 12-20 min were allowed between doses of the purinoceptor agonists, which were left in contact with the tissue for 2-5 min until a maximal relaxation had been observed.
Degradation studies Segments of rat colon longitudinal muscle (approximately 30 mg wet weight) were suspended in 3 ml organ baths as described for the pharmacological studies. Following preincubation for 90-120 min with frequent washing, each was exposed to ATP, AMPPCP and AMPCPP (100 jIM at 30 min intervals), and samples of the bathing medium were taken at 2, 5, 10 and 20 min and frozen for later analysis by high performance liquid chromatography as described by Welford et al. (1986, 1987).
PURINES AND THE RAT COLON LONGITUDINAL MUSCLE
887
Materials ATP, AMPPCP, AMPCPP, AMPCP, adenosine, ADA (Type VI) and NBTG were obtained from Sigma UK. Ltd, DPCPX, 2-MeSATP, CPA and NECA were obtained from Research Biochemicals, and suramin was a generous gift from Bayer, UK. Other purine analogues were kindly provided by Dr Noel J. Cusack, Whitby Research Inc, Richmond, Virginia, U.S.A. CPA (10 mM) was dissolved in 20% ethanol, DPCPX (1 mM) was dissolved in 2% aqueous dimethylsulphoxide (DMSO) containing 6 mM NaOH and NBTG (50 mM) was dissolved in DMSO. After dilution corresponding to the final bath concentration of the substances used these solvents had no effect on the responses of the tissue. The ADA solution was supplied in 50% glycerol0.01 M potassium phosphate and was diluted in distilled water to give a stock solution of 200 units ml-'.
t~0 40-I co
0 0.1
1
10
100
Purine concentration (>M)
Results Adenosine, ATP, AMPPCP, AMPCPP, NECA and CPA each relaxed the carbachol-contracted rat colon longitudinal muscle, and the order of potency was NECA> CPA = AMPCPP> adenosine = AMPPCP = ATP (Figure 1). DPCPX
d
a
100
Figure I Relaxation of the rat colon longitudinal muscle by adenosine (@), NECA (A), CPA (-), ATP (0), AMPPCP (') or AMPCPP (V). Each point is the mean of at least 15 determinations, and the error bars, which never exceeded 5%, have been omitted for clarity. For abbreviations, see text.
100 -
-
50 -
50 -
O-
0- l
10
1
100
100
ATP concentration (>.M)
Adenosine concentration (>M) 1
b
100
100 -i
-
50-
(a
x
la
W-J
0.1
1
1
10
100
AMPPCP concentration (>M)
NECA concentration (>M) f
C
100 -
100 -
50 -
50-
0-
10
I
I
I
1
10
100
CPA concentration (>±M)
0-
10
100
AMPCPP concentration (>M)
Figure 2 Relaxation of the rat colon longitudinal muscle by (a) adenosine, (b) NECA, (c) CPA, (d) ATP, (e) AMPPCP or (f) AMPCPP alone (@) or in the presence of DPCPX (3 JiM) (0). Each point is the mean of at least 4 determinations and the vertical bars show s.e.mean. For abbreviations, see text.
888
ST BAILEY & S.M.O. HOURANI
(3 gM) inhibited the responses to NECA, CPA, adenosine, AMPPCP and ATP, but did not affect the relaxations induced by AMPCPP (Figure 2), whereas suramin (3001MM) inhibited the responses to AMPCPP but not those of the other agonists (Figure 3). DPCPX (1 gM) caused smaller shifts to the right of the log concentration-response curves to all the agonists except for AMPCPP (results not shown), but Schild analysis was not performed as concentrations of DPCPX above 3pM could not be achieved. L-AMPPCP (100 gM) and 2-MeSATP (100m M) were almost inactive, ADP-P-F(l00 Mm) induced a relaxation similar to that of ATP (100 IM), while adenosine 5'-(Q,-methylene) diphosphonate (AMPCP) (1001MM) induced a relaxation similar to that of AMPCPP (100m M) (Figure 4). Representative traces showing the relaxations induced by the agonists are shown in Figure 4. The responses to AMPCP were inhibited by suramin (300 gM) in a similar manner to those of AMPCPP (results not shown). Incubation of the tissues with ADA (2 units ml-') abolished the relaxations induced by adenosine and inhibited those induced by ATP and AMPPCP, but had no effect on those induced by AMPCPP (Figure 5). Incubation of the tissues with NBTG (50 gM) potentiated responses to adenosine, causing a roughly three fold shift to the left of the log concentration-response curve, and also potentiated responses to AMPCPP but to a somewhat lesser extent (results not shown).
Degradation studies Dephosphorylation of ATP by the rat longitudinal muscle preparation was rapid, with approximately 80% being degraded, largely to adenosine and AMP, during 30 min incubation. AMPPCP was degraded more slowly to adenosine, with approximately 20% conversion during 30 min incubation, and AMPCPP was degraded to AMPCP, with approximately 30% conversion over 30 min (Figure 6).
Discussion These results show that the outer longitudinal muscle layer of the rat colon contains P1- and P2-purinoceptors which both mediate relaxation, and that the responses of this tissue to purines are qualitatively similar to the responses seen in the guinea-pig taenia caeci, the longitudinal muscle of the guinea-pig caecum. For the Pl-purinoceptor in the longitudinal muscle the potency order of agonists is NECA> CPA> adenosine, and DPCPX has an apparent dissociation constant in the micromolar range, suggesting that the receptor is of the A2 subtype (Bruns, 1990; Collis, 1990; Daly, 1990). DPCPX (3 gM) caused a somewhat smaller shift to the right of the log concentration-response curve to adenosine compared with NECA and CPA, which may suggest that it also interfered with adenosine uptake. In general the postd
a
100-
100
50
50-
-
0.
0 *1
10
100
1
100
ATP concentration (>tM)
Adenosine concentration (>xM) 100 -1
e 100-
50 -
50-
b
10
R C
-
0
ur
(U
CC
0-
0-
0.1
1
1
10
100
AMPPCP concentration (>tM)
NECA concentration ({M) c
10
f
100-
100-
50-
50-
0-
0 10
100
CPA concentration (>M)
1
10
100
AMPCPP concentration (>M )
Figure 3 Relaxation of the rat colon longitudinal muscle by (a) adenosine, (b) NECA, (c) CPA, (d) ATP, (e) AMPPCP or (f) AMPCPP alone (@) or in the presence of suramin (300 IpM) (0). Each point is the mean of at least 4 determinations and the vertical bars show s.e.mean. For abbreviations, see text.
PURINES AND THE RAT COLON LONGITUDINAL MUSCLE
C Adenosine
C NECA
C CPA
J1 g
/t
4 min
r
C ATP
C AMPPCP
C AMPCPP
C AMPCP
C 2t C 2-MeSATP
C ADP-3-F
C L-AMPPCP
Figure 4 Representative traces showing relaxations of the rat colon longitudinal muscle to purines after precontraction of the tissue with carbachol (1 #tM) (C). The concentration of purine added in each case was 100 IAM, except for NECA which was 10 I1M. For abbreviations used in figure, see text.
100-
50 -
50-
0-
0C
synaptic inhibitory effects of adenosine in smooth muscle including the guinea-pig taenia caeci are mediated by A2 receptors, so the finding that the adenosine receptor here is of the A2 subtype is consistent with results in other tissues (White, 1988; Stone, 1989; Kennedy, 1990; Olsson & Pearson, 1990). In those tissues in which adenosine causes contraction, such as the renal vasculature (Kenakin & Pike, 1987) and the guinea-pig myometrium (Smith et al., 1988), this appears to be via Al receptors, and this is also the case in the rat colon muscularis mucosae (Bailey et al., 1992). It has also been reported that the guinea-pig aorta and trachea each possess both Al and A2 receptors which mediate contraction and relaxation respectively, although for adenosine the A2-mediated relaxation is dominant (Farmer et al., 1988; Stoggall & Shaw, 1990). The existence in the rat colon of excitatory Al receptors (on the muscularis mucosae) and inhibitory A2 receptors (on the longitudinal muscle) is therefore consistent with this general pattern, although the two receptor types are clearly at anatomically distinct locations within the colon, a possibility which has not been investigated in the other tissues. The rat colon longitudinal muscle also possesses P2purinoceptors, as AMPCPP caused a relaxation which was inhibited by the P2-purinoceptor antagonist suramin but not by DPCPX. The actions of ATP and AMPPCP appeared however to be mediated via the A2 receptor, as they were not inhibited by suramin but were inhibited by DPCPX at the same concentration as was adenosine. It seems likely that this A2 effect was at least partially indirect and a result of the degradation of ATP and AMPPCP to adenosine, as the responses to ATP and AMPPCP were reduced by ADA, which had no effect on the responses to AMPCPP. The adenosine uptake inhibitor NBTG potentiated responses to adenosine as expected, but also potentiated responses to AMPCPP which does not act on A2 receptors here, and therefore was not useful for resolving the question as to whether the actions of ATP and AMPPCP were direct or via adenosine. The potentiation by NBTG of responses to AMP-
b 100 -j
a
1
0
100
1
100 Adenosine concentration (j±M) 10
ATP concentration (jiM)
x
cc
d
c
100-
100 -,
50-
50-
0-
01
10
100
AMPPCP concentrtion (gM)
889
1
10
100
AMPCPP concentration
(gM)
Figure 5 Relaxation of the rat colon longitudinal muscle by (a) adenosine, (b) ATP, (c) AMPPCP or (d) AMPCPP alone (0) or in the presence of adenosine deaminase (2 units ml- 1) (0). Each point is the mean of at least 5 determinations and the vertical bars show s.e.mean. For abbreviations, see text.
890
S.J. BAILEY & S.M.O. HOURANI 100
a -k
b
100 -
0
ci
C1)
50
C) 0 C.)
a1)
C)
CL
0
T
10
0
20
7 30
100 c
50-
0
the 0
10
I
I
20
30
Incubation time (min)
Figure 6 Degradation of purines by the rat colon longitudinal muscle preparation. Removal of (a) ATP, (b) AMPPCP or (c) AMPCPP (0) and appearance of ADP (A), AMP (0), adenosine (V) or AMPCP (0). Each point is the mean of at least 4 determinations and the vertical bars show s.e. mean. For abbreviations, see text.
CPP was presumably due to increased endogenous adenosine levels enhancing all relaxant responses, but this was not investigated further. It is probable that AMPCPP retained its
P2-purinoceptor activity even though it is also significantly degraded, to at least the same extent as AMPPCP, because its breakdown product was not adenosine but AMPCP, which was inhibited by suramin and therefore acts on P2purinoceptors. These findings are in contrast to our results with the rat colon muscularis mucosae and the rat duodenum, where it is clear that despite its rapid degrada-
tion, ATP has little if any P1-purinoceptor activity whereas AMPPCP acts almost entirely and directly via P1purinoceptors (Bailey & Hourani, 1990; Hourani et al., 1991). Unfortunately the uncertainty regarding the direct A2 receptor activity of AMPPCP itself in the longitudinal muscle means that the use of this tissue has not resolved our initial hypothesis that AMPPCP might have Al-selective PIpurinoceptor activity, as had been suggested by our previous studies (Bailey & Hourani, 1990; Nicholls et al., 1990; Hourani et al., 1991; Bailey et al., 1992; Nicholls et al., 1992). Although the P2-purinoceptor which mediates relaxation in the guinea-pig taenia caeci is clearly of the P2Y subtype, with an order of potency of 2-MeSATP> ATP> AMPCPP (Burnstock et al., 1984; Burnstock & Kennedy, 1985), the P2purinoceptor in the rat colon longitudinal muscle is not so easy to define by use of these agonists as ATP itself had no P2-purinoceptor activity in this tissue and 2-MeSATP was almost inactive. The weak activity of 2-MeSATP may be due to its rapid breakdown to 2-methylthioadenosine, which is somewhat less active than adenosine on the A2 receptors in the guinea-pig taenia caeci (Satchell & Maguire, 1975) and is therefore likely to be less active in the longitudinal muscle preparation. Indeed the very rapid degradation of the nucleotides, including the normally resistant AMPPCP and AMPCPP, by this preparation makes analysis of the structure-activity relationships of the P2-purinoceptor very difficult. The response mediated by this receptor is a relaxation which is qualitatively similar to that seen in the guineapig taenia caeci, which might suggest involvement of a P2ypurinoceptor, but this is not a reliable method for receptor classification, particularly in view of the existence of P2Ypurinoceptors which mediate contraction on the muscularis mucosae and the neonatal rat duodenum (Bailey & Hourani, 1990; Nicholls et al., 1990). The lack of activity of LAMPPCP and the activity of ADP-P-F in the longitudinal muscle would appear to confirm that the P2-purinoceptor here is of the P2y subtype, as L-AMPPCP and ADP-13-F are selective agonists at P2X- and P2Y-purinoceptors respectively (Hourani et al., 1986; 1988). However, the same argument could in theory apply to L-AMPPCP as to 2-MeSATP, as its potential breakdown product, L-adenosine, is inactive on PIpurinoceptors (Brown et al., 1982), and similarly ADP-P-F could be acting via its probable breakdown product, adenosine, so this evidence cannot be regarded as conclusive. What is needed, both to define the P2-purinoceptor in this tissue and to resolve the question of whether ATP and AMPPCP act directly or indirectly on the P,-purinoceptor, is a selective inhibitor of the ectonucleotidases which are responsible for the breakdown. Notwithstanding these uncertainties, it is clear from the work presented here together with our previous work (Bailey & Hourani, 1990; Bailey et al., 1992), that the two preparations from the rat colon, the muscularis mucosae and the longitudinal muscle, have different P,-purinoceptor subtypes, Al mediating contraction and A2 mediating relaxation, and that they also possess P2 purinoceptors which mediate the same effect in each case as the P1 purinoceptors. The physiological significance of these receptors is unknown, but probably reflects the widespread roles of purines as transmitters and modulators in the autonomic nervous system. The nature of the P2-purinoceptor on the longitudinal muscle is still undefined, but it is likely to be of the P2Y subtype as in the muscularis mucosae, although the response observed is different in the two preparations. In the longitudinal muscle preparation the breakdown of nucleotides is extremely rapid, and is considerably faster than that previously observed in the muscularis mucosae in spite of the fact that the longitudinal muscle is a much smaller tissue, with a wet weight of about 30 mg compared to around 200 mg for the muscularis mucosae. This very rapid degradation undoubtedly affects the responses to the nucleotides, and this makes the study of the pharmacology of the P2-purinoceptors in this
PURINES AND THE RAT COLON LONGITUDINAL MUSCLE
tissue somewhat problematical, and shows clearly the care which must be taken when attempting to use nucleotide agonist potencies to define receptor subtypes.
891
We thank the Medical Research Council for support for S.J.B. (G8723473N) and Dr Noel Cusack for supplying analogues.
References
BAILEY, S.J. & JORDAN, C.C. (1984). A study of [D-Pro2, D-Phe7 D-Trpl-substance P and (D-Trp7f9l-substance P as tachykinin partial agonists in the rat colon. Br. J. Pharmacol., 82, 441-451. BAILEY, S.J., HICKMAN, D. & HOURANI, S.M.O. (1992). Characterisation of the PI-purinoceptors mediating contraction of the rat colon muscularis mucosae. Br. J. Pharmacol., 105, 400-404. BAILEY, S.J. & HOURANI, S.M.O. (1990). A study of the purinoceptors mediating contraction in the rat colon. Br. J. Pharmacol., 100, 753-756. BROWN, C., BURNSTOCK, G., CUSACK, N.J., MEGHJI, P. & MOODY,
C.J. (1982). Evidence for the stereospecificity of the PIpurinoceptor. Br. J. Pharmacol., 75, 101-107. BRUNS, R.F. (1990). Adenosine receptors: roles and pharmacology.
Ann N.Y. Acad. Sci., 603, 211-226. BRUNS, R.F., LU, G.H. & PUGSLEY, T.A. (1986). Characterization of the A2 adenosine receptor labeled by [3H]NECA in rat striatal membranes. Mol. Pharmacol., 29, 331-346. BURNSTOCK, G. (1978). A basis for distinguishing two types of purinergic receptor. In Cell and Membrane Receptors for Drugs and Hormones: a Multidisciplinary Approach. ed. Bolis, L. & Straub, R.W. pp. 107-118. New York: Raven Press. BURNSTOCK, G. (1990). Classification and characterization of purinoceptors. In Purines in Cellular Signalling. ed. Jacobson, K.A., Daly, J.W. & Manganiello, V., pp. 241-253. New York:
Springer-Verlag.
BURNSTOCK, G., HILLS, J.M. & HOYLE, C.H.V. (1984). Evidence that the PI-purinoceptor in the guinea-pig taenia coli is an A2 subtype. Br. J. Pharmacol., 81, 533-541. BURNSTOCK, G. & KENNEDY, C. (1985). Is there a basis for distinguishing two types of P2-purinoceptor? Gen. Pharmacol., 16, 433-440. COLLIS, M.G. (1990). Adenosine receptor sub-types in isolated tissues: antagonist studies. In Purines in Cellular Signalling. ed. Jacobson, K.A., Daly, J.W. & Manganiello, V. pp. 48-53. New York: Springer-Verlag. COLLIS, M.G., JACOBSON, K.A. & TOMKINS, D.M. (1987). Apparent affinity of some 8-phenyl-substituted xanthines at adenosine receptors in guinea-pig aorta and atria. Br. J. Pharmacol., 92, 69-75. COLLIS, M.G. & PETTINGER, S.J. (1982). Can ATP stimulate PIreceptors in guinea-pig atrium without conversion to adenosine? Eur. J. Pharmacol., 81, 521-529. COLLIS, M.G., STOGGALL, S.M. & MARTIN, F.M. (1989). Apparent affinity of 1,3-dipropyl-8-cyclopentylxanthine for adenosine Al and A2 receptors in isolated tissues from guinea-pigs. Br. J. Pharmacol., 97, 1274-1278. CUSACK, N.J. & HOURANI, S.M.O. (1990). Subtypes of P2purinoceptors. Studies using analogues of ATP. Ann. N.Y. Acad. Sci., 603, 172-181. CUSACK, N.J., HOURANI, S.M.O., LOIZOU, G.D. & WELFORD, L.A. (1987). Pharmacological effects of isopolar phosphonate analogues of ATP on P2-purinoceptors in guinea-pig taenia coli and urinary bladder. Br. J. Pharmacol., 90, 791-795. DAHLEN, S.-E. & HEDQVIST, P. (1980). ATP, P,y-methylene-ATP, and adenosine inhibit non-cholinergic non-adrenergic transmission in rat urinary bladder. Acta Physiol, Scand., 109, 137-142. DALY, J.W. (1990). Adenosine agonists and antagonists. In Purines in Cellular Signalling. ed. Jacobson, K.A., Daly, J.W. & Manganiello, V. pp. 3-12. New York: Springer-Verlag. DEN HERTOG, A., NELEMANS, A. & VAN DEN AKKER, J. (1989a). The inhibitory action of suramin on the P2-purinoceptor in smooth muscle cells of guinea-pig taenia caeci. Eur. J. Pharmacol., 166, 531-534. DEN HERTOG, A., VAN DEN AKKER, J. & NELEMANS, A. (1989b). Suramin and the inhibitory junction potential in taenia caeci of the guinea-pig. Eur. J. Pharmacol., 173, 207-209. DUNN, P.M. & BLAKELY, A.G.H. (1988). Suramin: a reversible P2purinoceptor antagonist in the niouse vas deferens. Br. J. Pharmacol., 93, 243-245. FARMER, S.G., CANNING, B.J. & WILKINS, D.E. (1988). Adenosine receptor-mediated contraction and relaxation of guinea-pig isolated tracheal smooth muscle: effects of adenosine antagonists. Br. J. Pharmacol., 95, 371-378.
GORDON, J.L. (1986). Extracellular ATP: effects, sources and fate. Biochem. J., 233, 309-319. HOURANI, S.M.O., BAILEY, S.J., NICHOLLS, J. & KITCHEN, I. (1991).
Direct effects of adenylyl 5'-(P,y-methylene) diphosphonate, a stable ATP analogue, on relaxant PI-purinoceptors in smooth muscle. Br. J. Pharmacol., 104, 685-690. HOURANI, S.M.O. & CHOWN, J.A. (1989). The effects of some possible inhibitors of ectonucleotidases on the breakdown and pharmacological effects of ATP in the guinea-pig urinary bladder. Gen. Pharmacol., 20, 413-416. HOURANI, S.M.O., LOIZOU, G.D. & CUSACK, N.J. (1986). Pharmacological effects of L-AMP-PCP on ATP receptors in smooth muscle. Eur. J. Pharmacol., 131, 99-103. HOURANI, S.M.O., WELFORD, L.A., LOIZOU, G.D. & CUSACK, N.J.
(1988). Adenosine 5'-(2-fluorodiphosphate) is a selective agonist at P2-purinoceptors mediating relaxation of smooth muscle. Eur. J. Pharmacol., 147, 131-136. HOYLE, C.H.V., KNIGHT, G.E. & BURNSTOCK, G. (1990). Suramin antagonizes responses to P2-purinoceptor agonists and purinergic nerve stimulation in the guinea-pig urinary bladder and taenia coli. Br. J. Pharmacol., 99, 617-621. JACOBSON, K.A. (1990). Molecular probes for adenosine receptors. In Purines in Cellular Signalling. ed. Jacobson, K.A., Daly, J.W. & Manganiello, V., pp. 54-64. New York: Springer-Verlag. KENAKIN, T.P. & PIKE, N.B. (1987). An in vitro analysis of purinemediated renal vasoconstriction in rat isolated kidney. Br. J. Pharmacol., 90, 373-381. KENNEDY, C. (1990). PI- and P2-purinoceptor subtypes - an update. Arch. Int. Pharmacodyn., 303, 30-50. LEFF, P., WOOD, B.E. & O'CONNOR, S.E. (1990). Suramin is a slowlyequilibrating but competitive antagonist at P2X-receptors in the rabbit isolated ear artery. Br. J. Pharmacol., 101, 645-649. MOODY, C.J. & BURNSTOCK, G. (1982). Evidence for the presence of P1-purinoceptors on cholinergic nerve terminals in the guinea-pig ileum. Eur. J. Pharmacol., 77, 1-9. NICHOLLS, J., HOURANI, S.M.O. & KITCHEN, I. (1990).- The ontogeny of purinoceptors in rat urinary bladder and duodenum. Br. J. Pharmacol., 100, 874-878. NICHOLLS, J., HOURANI, S.M.O. & KITCHEN, I. (1992). Characterization of P1-purinoceptors on rat duodenum and urinary bladder. Br. J. Pharmacol., (in press). OLSSON, R.A. & PEARSON, J.D. (1990). Cardiovascular purinoceptors. Pharmacol. Rev., 70, 761-845. PEARSON, J.D. & SLAKEY, L.L. (1990). Intracellular and extracellular metabolism of adenosine and adenine nucleotides. In Purines in Cellular Signalling. ed. Jacobson, K.A., Daly, J.W. & Manganiello, V. pp. 13-19. New York: Springer-Velag. ROMANO, C. (1981). Nicotine action on rat colon. J. Pharmacol. Exp. Ther., 217, 828-833. ROSE'MEYER, R.B. & HOPE, W. (1990). Evidence that A2 purinoceptors are involved in endothelium-dependent relaxation of the rat thoracic aorta. Br. J. Pharmacol., 100, 576-580. SATCHELL, D.G. & MAGUIRE, M.H. (1975). Inhibitory effects of adenine nucleotide analogs on the isolated guinea-pig taenia coli. J. Pharmacol. Exp. Ther., 195, 540-548. SHINOZUKA, K., BJUR, R.A. & WESTFALL, D.P. (1988). Characterisation of prejunctional purinoceptors on adrenergic nerves of the rat caudal artery. Naunyn-Schmiedebergs Arch. Pharmacol., 338, 221-227. SMITH, M.A., BUXTON, I.L.O. & WESTFALL, D.P. (1988). Pharmacological classification of receptors for adenyl purines in guinea-pig myometrium. J. Pharmacol. Exp. Ther., 247, 10591063. STOGGALL, S.M. & SHAW, J.S. (1990). The coexistence of adenosine Al and A2 receptors in guinea-pig aorta. Eur. J. Pharmacol., 190, 329-335. STONE, T.W. (1989). Purine receptors and their pharmacological roles. Adv. Drug Res., 18, 291-429. WELFORD, L.A., CUSACK, N.J. & HOURANI, S.M.O. (1986). ATP analogues and the guinea-pig taenia coli: a comparison of the structure-activity relationships of ectonucleotidases with those of the P2-purinoceptor. Eur. J. Pharmacol., 129, 217-224.
892
S.J. BAILEY & S.M.O. HOURANI
WELFORD, L.A., CUSACK, N.J. & HOURANI, S.M.O. (1987). The
structure-activity relationships of ectonucleotidases and of excitatory P2-purinoceptors: evidence that dephosphorylation of ATP analogues reduces pharmacological potency. Eur. J. Pharmacol., 141, 123-130. WHITE, T.D. (1988). Role of adenine compounds in autonomic neurotransmission. Pharmacol. Ther., 38, 129-168.
WIKLUND, N.P., GUSTAFSSON, L.E. & LUNDIN, J. (1985). Pre- and post-junctional modulation of cholinergic neuroeffector transmission by adenine nucleotides. Experiments with agonist and antagonist. Acta Physiol. Scand., 125, 681-691. (Received October 3, 1991 Revised December 6, 1991 Accepted December 18, 1991)
